This is a new collaborative project started in early 1997. The specific aim is to our advance knowledge of how the brain represents language; we will integrate spatial and temporal information obtained by MEG and fMRI during human brain processing of linguistic stimuli. Specifically, we will give simple linguistic and nonlinguistic stimuli to subjects, and use fMRI and MEG to map sensory (primary and association) and language areas activated by the stimuli. Using fMRI to identify areas of high activity, spatial information can be used as possible source sites for dipole calculations in analyzing MEG data. So far we have collected data on sensory stimuli with good initial success.